Skip to content
Sign inRequest access

Products

GPU CloudAMD MI325X and NVIDIA H200 by the hour or on reservation.
Dedicated ClustersPreviewPrivate, interconnected GPU clusters for sustained workloads.
SandboxesPreviewIsolated, short-lived environments for developers and AI agents.
Serverless GPUPreviewGPU containers that scale with requests, down to zero.
Serverless InferenceOpen models behind one OpenAI-compatible endpoint.
Dedicated EndpointsReserved serving capacity for one model, billed per GPU-hour.
PlaygroundSend a prompt to any model your workspace can reach.
Model DeploymentsPreviewImmutable deployments with instant promote and rollback.
Fine-tuningLoRA fine-tunes on the open models in our catalog.
Model TrainingPreviewFoundation-model training runs, managed end to end.
EvaluationsPreviewBenchmark suites and model comparisons you can rerun.
DatasetsManaged and synthetic training data, versioned and ready to use.
Data ManagementPreviewDatasets, connections and storage, with lineage across the platform.
PipelinesPreviewVersioned DAGs that automate data, training and deployment.
Data RoomsPreviewSecure document sharing with guests, NDAs and an audit trail.
NokuvaPreviewThe AI-native design editor.
TavocPreviewA Rust-native code platform: IDE and desktop apps.
FoltracComing soonMulti-cloud deployment with autonomous operations.
W0rkTreeComing soonVersion control for agent fleets.
Identity & AccessMembers, roles, policies, SSO and audit for every workspace.
Billing & CostUsage, invoices, credits and budgets in one place.
Logs & MonitoringRequest logs, latency and error rates across services.
Project CanvasPreviewSee and change a project's resources on one live canvas.

Pages

Home/
Platform/platform
All products/products
Pricing/pricing
GPU pricing/pricing#compute
Inference services/services/inference
Model training services/services/model-training
Protocol design/services/protocol-design
Research partnerships/services/research-partnerships
Protocols/protocols
Datasets/datasets
Documentationdocs.legionedge.ai
Changelog/changelog
News/news
Research/research
Blog/blog
Articles/articles
Events/events
Startup Program/programs/startup
Ambassador Program/programs/ambassador
Partner Program/programs/partner
Event Program/programs/events
Company/company
Careers/careers
Partners/partners
Contact/contact
Statusstatus.legionedge.ai
Terms of Service/legal/terms
Privacy Policy/legal/privacy

Model Deployments

Preview

Ship models the way you ship code.

Group a model's releases into a project: every deployment is immutable, production is an alias you promote or roll back instantly, and each project has its own domains, environment variables, logs and analytics. Serve it managed on shared capacity or dedicated on your own GPUs.

Request accessView pricing
Pricing
Billed as inference or per GPU-hour
Layer
Inference
Deploy from
Catalog models, Fine-tune checkpoints, Hugging Face, Uploaded weights, Pipeline outputs, vLLM, SGLang, TGI

Why Model Deployments

Every release is a deployment you can return to.

A model project keeps every deployment it has ever made. Nothing is edited in place: a new config or checkpoint is a new deployment, and production is an alias that points at one of them.

  • Managed or dedicated

    Shared, autoscaled capacity or GPUs reserved for you.

  • Instant rollback

    Production is an alias: move it back in one step.

  • Logs and analytics

    Requests, latency and errors for every deployment.

How it works

Immutable deployments with instant promote and rollback.

  1. Pick a source

    A catalog model, a fine-tune checkpoint, a Hugging Face repo and revision, uploaded weights or a pipeline output.

  2. Choose managed or dedicated

    Managed runs on shared, autoscaled capacity. Dedicated runs on your GPUs, with the hardware, regions and replicas you pick.

  3. Watch it build

    Resolve the artifact, pull weights, quantize, load on GPU, warm up and pass a health check, with each phase’s logs streaming live.

  4. Promote or roll back

    Test a staging or preview deployment on its own URL, promote it to production, and roll back in one step if it misbehaves.

In the console

Everything in one place.

Model Deployments lives in the LegionEdge cloud console, next to every other service in the workspace, under one account and one identity.

  • Project overview

    The production deployment up front, with its domains, source, capacity and the actions you need.

    • Instant Rollback, Promote and Redeploy
    • Preview and staging deployments listed below
    • Who deployed it, and from which pipeline run

  • Deployment detail

    Everything about one immutable deployment, from its build to the replicas serving it.

    • Build timeline with streaming logs per phase
    • Frozen config, diffed against production
    • A playground that calls this deployment’s own URL

  • Analytics and logs

    Requests, latency, time to first token, tokens, errors and cost, broken down by deployment, region and key.

    • Sampled, searchable request logs with live tail
    • Prompt and response logging off by default
    • Opt in per project, with a retention setting

  • Settings

    The controls a production endpoint needs, managed per project.

    • Custom domains, verified with a TXT record
    • Environment variables and secret references
    • Firewall, per-key rate limits and guardrails

Pricing

Billed as inference or per GPU-hour

Managed deployments are billed per token, like serverless inference. Dedicated deployments are billed per GPU-hour for the capacity they run on.

  • The new-project wizard shows a cost estimate before you deploy.
  • Scale-to-zero is optional on dedicated deployments.
Full pricing

Catalog models

Catalog models with context window and token prices
ModelContextInput / 1MOutput / 1M
DeepSeek V4 Flashdeepseek-v4-flash160K$0.30$1.20
DeepSeek V4 Prodeepseek-v4-pro800K$1.74$3.48
Qwen3 0.6Bqwen3-0.6b32K$0.10$0.20
Qwen3 4B Instructqwen3-4b-instruct32K$0.20$0.60
Qwen3.6 27Bqwen3-6-27b256K$0.30$0.90
Qwen3.8 27Bqwen3-8-27b256K$0.30$0.90
GLM-5.2glm-5-2128K$0.50$2.00
GLM-5.3glm-5-31M$0.50$2.00
Kimi K3kimi-k31M$0.60$2.50
Kimi K3 Turbokimi-k3-turbo128K$0.60$2.50

GPU rates

GPU rates in USD per GPU-hour, on demand and by reservation term
GPUMemoryOn demand1-month6-month12-month
AMD Instinct MI325X256 GB HBM3E$4.79$4.31$3.83$3.35
NVIDIA H200141 GB HBM3e$4.79On request

Related products

Pairs well with Model Deployments

Get started

Start with a workload.

Connect to the platform directly, or work with our team on a specific technical challenge.

Self-service

LegionEdge Cloud

Create a workspace, explore the available models, and connect your application. The documentation takes you through your first request.

  • Turn on only the services you need
  • Pay per token or per GPU-hour
  • One account, one bill
Create a workspaceRead the docs

Custom engagements

Work with our team

Bring your data requirements, model goals, or infrastructure constraints. We'll define the scope, deliverables, and next steps together.

  • Scoped with an evaluation baseline
  • Dedicated clusters and reserved GPUs
  • A named engineer from day one
Discuss your projectsales@legionedge.ai